We found that rhabdomyosarcomas show consistently high levels of TGF- beta1, and to a lesser extent TGF-beta3, but lack TGF-beta2 protein by immunohistochemical staining. We also detected variable levels of TGF- beta protein synthesis in conditioned media of cultured rhabdomyosarcoma (RMS) cells using a bioassay based on TGF-beta inhibition of DNA synthesis in mink lung fibroblasts (CCL-64 cells). Furthermore, all RMS expressed variable levels of TGF-beta1 mRNA in vitro. These data, in combination with our previous data of TGF-beta- induced inhibition of RMS cell differentiation in vitro, and the known inhibitory effect of TGF-beta in normal myogenesis, have suggested to us a possible autocrine inhibitory role of TGF-beta in human RMS. This hypothesis will be studied in the following ways: (1) RMS cell lines expressing high and low levels of TGF-beta mRNA will be studied for the presence of TGF-beta receptors by PAGE of cell lysates incubated with (125)I-TGF-beta 1,2,3. (2) Anti-TGF-beta blocking antibodies and antisense oligonucleotides will be employed to investigate interference with cell growth (DNA synthesis by 3H-thymidine incorporation) and inhibition of myogenic differentiation (myotube fusion assay), in the presence or absence of increased concentration of exogenous TGF-beta. Since TGF-beta receptors decrease after fusion of myoblasts into myotubes, and our previous data have shown a more potent action of TGF- beta in cell lines with a more primitive morphology, we will also evaluate levels of TGF-beta mRNA and protein in fusion arrested (with diazepam or EGTA) , versus 5-azacytidine differentiated RMS cells. Retinoic acid (RA)-treated cells will be evaluated as well, since RA was found to induce TGF-beta receptor expression in some cell types.